The Safe Drinking Water Act (SDWA) mandates that municipal water utilities monitor their water. Under the SDWA, the number of monitoring sites in the outgoing distribution system depends upon the number of customers served by the utility. For example, in large utilities serving more than 100,000 customers, the utilities must provide monitoring at 100 sites in the distribution system.
In the drinking water filtration industry, common contaminants of concern are trihalomethanes, biological contamination, nitrate and heavy metals, such as lead. Each are removed by different means within the filtration system. Trihalomethanes are effectively removed with charcoal. Biological contamination such as cysts are removed with fine mesh mechanical filtration. Nitrate and lead are removed by either of two methods, reverse osmosis (RO) or ion exchange resins.
RO systems are effective for removing nitrates and heavy metals. Most quality systems offer a monitor that indicates that there is a rupture in the RO membrane and thus the system requires membrane replacement. Such monitors generally measure the conductivity of the input water and the output water. When the membrane is intact, the conductivity of the input water will differ from that of the output water to the extent that the system is removing inorganic contaminants from the water. When the membrane ruptures, allowing the input water to flow through the membrane, the difference between the conductivity of the input water and the output water will lessen beyond a pre-set threshold and trigger a signal to the user. A disadvantage to RO systems is that they require about five gallons of water to back-flush the membrane for every filtered gallon available for use. For a typical system delivering five gallons of water per day, an RO system will use up to 25 gallons of water per day to back-flush. Thus, while effective for removing inorganic contaminants, RO systems are very wasteful of water.
Ion exchange resins come from the manufacturer in the form of beads, having ion exchange sites on the beads. A plurality of such beads are typically agglomerated together to form a resin bed. Cation resins commonly have sodium on the exchange sites and anion resins commonly have chloride ions on the exchange sites. In ion exchange resins, heavier ions displace lighter ions. The following table sets forth a list of cations together with their selectivity coefficients. Selectivity coefficients are indicators of the preference of the resin for each of the ions relative to hydrogen.
TABLE 1 ______________________________________ Cation Selectivity Coefficients of Four Cation Resins selectivity coefficients Cross-linking, wt % Ion symbol 4 8 12 16 ______________________________________ hydrogen H 1.0 1.0 1.0 1.0 iron Fe 2.4 2.55 2.7 2.9 zinc Zn 2.6 2.7 2.8 3.0 cadmium Cd 2.8 2.95 3.3 3.95 calcium Ca 3.4 3.9 4.6 5.8 strontium Sr 3.9 4.95 6.25 8.1 copper Cu 3.2 5.3 9.5 14.5 mercury Hg 5.1 7.2 9.7 14.0 lead Pb 5.4 7.5 10.1 14.5 ______________________________________
As can be seen from the chart, cations such as iron (Fe), zinc (Zn) and calcium (Ca) have lower preference ratings than mercury (Hg) and lead (Pb). For example, in a cation resin bed having Ca on the exchange sites, if Hg were introduced into the bed, the Hg ions would displace the Ca ions, since Hg is more highly preferred by the ion exchange resin than Ca. If Pb were subsequently introduced, the Pb would displace lighter ions on the resin, and so on.
There are some batch on-line analyzers available on the market that can detect and quantify the presence of contaminants. However, each of these systems is extremely expensive. One system, ChemScan Process Analyzers, available from Applied Spectrometry Associates, Inc. of Waukesha, Wis., uses ultraviolet-visible spectrometry to detect contaminants. This analyzer costs in the $20,000-$40,000 range depending on the contaminants being detected. Ionics, Inc. of Watertown, Mass. offers the OVA 3000 series Trace Chemical Analyzers using the Wet Chemical method for lighter metals and Anodic Stripping Voltammnetry for heavier metals. Those systems cost about $40,000. For a large system, having 100 sites, the capital cost of installing such systems would be $4,000,000, which, for a utility serving 100,000 customers, would effect a $40 per customer one time charge for water monitoring. Thus, there is a need for a continuous, on-line system that is economical.
Though water is monitored when it leaves the municipal water plant, some contaminants may get into the water before the water is dispensed from the household tap. One contaminant of special note, lead, is highly toxic. It is present in lead solder in household plumbing, sometimes in the plumbing itself and sometimes in the water's delivery system. Water filtration systems that rely on cation exchange resin technology to remove lead or other toxic heavy metals can work effectively until the ion exchange system is no longer able to capture all of the heavy metals. This point is called the break-through stage. Therefore, it is necessary to detect when the break-through stage is reached. However, the monitor used to detect rupture in an RO membrane will not work in this application as exchange resins saturate gradually with no clearly detectable event such as occurs when an RO membrane ruptures. Thus, there is a need for a monitor to detect the presence of a specific ion, which is a threshold ion in a water filtration system.
Referring to the table, the logical stage to detect cation breakthrough in water from municipal water systems is at the copper level, having the effect of maximizing the longevity of the filtration cartridge and minimizing the health risks. However, an earlier stage threshold ion of cation breakthrough, such as zinc, is preferred for well water users to protect from such harmful ions as cadmium, which would be removed by municipal systems but may be present in wells.
Until recent years, standard anion exchange resins were used to remove nitrate from water. However, sulfates, which are common in nature, had higher selectivity coefficients than nitrate. The result was nitrate sloughing or dumping. That is, if an anion resin column was saturated with nitrate and sulfate was introduced into the column, the sulfates would displace the nitrates, thus, dumping the previously accumulated nitrates into the output water of a filter. Since nitrate has no taste, color or smell, the user was unaware of this event. To correct the problem, ion exchange resin manufacturers developed nitrate selective anion exchange resins which reversed the selectivity coefficients of nitrate and sulfates and thus the problem was solved. However, a disadvantage to using nitrate selective anion resins is that they are about 33% less efficient, i.e., the nitrate selective resins last about a third less long than a conventional anion exchange resin column. Thus, a monitor using nitrate as a threshold ion would allow the use of the more efficient conventional anion resin, maximize its longevity, and provide an alert to the user that the filter cartridge needed replacement.
In addition to the SDWA, the Clean Water Act (CWA) requires that wastewater treatment plants monitor the influent to their plants for specified contaminants. The CWA also specifies that industrial companies monitor their effluent that feeds into the wastewater stream. Such companies are referred to as Significant Industrial Users or SIU's. These SIU's will typically enter into pre-treatment agreements with their wastewater treatment plants covering the frequency of their monitoring requirement and the contaminants to be monitored.
The following table shows the Maximum Contaminant Levels (MCL's) in parts per million (ppm) of selected inorganic contaminants as mandated by the Environmental Protection Agency (EPA) under the Clean Water Act:
______________________________________ Contaminant Symbol MCL (ppm) ______________________________________ Copper Cu 1.3 Lead Pb 0.015 Zinc Zn 5.0 Mercury Hg 0.002 Arsenic As 0.050 ______________________________________
The frequency of monitoring of wastewater by both the wastewater treatment plants and the SIU's can be annually or more frequently. The common practice for monitoring is to collect water samples in the wastewater stream and submit those samples to testing laboratories for analysis. While the current practice is sufficient to comply with the mandates of the Clean Water Act, there is a possibility that a surge or spike of a contaminant can get into the wastewater treatment plant sludge undetected. Such an event may result in a costly cleanup process and possible fines from the EPA. Thus, there is a need for an in-line, continuous monitor that will detect and quantify contamination spikes in a water flow.
As shown in the Table above, zinc, having an MCL of 5 parts per million is different from mercury, having an MCL of 2 parts per billion. A concentration of 100 parts per billion of mercury would be a spike requiring immediate action. However, a similar reading of 100 parts per billion of zinc would be well within limits and require no action. Therefore, a monitor for heavy metals needs to be sensitive enough to measure vastly differing concentration levels for different contaminants.
As noted above, the USEPA mandates specific MCL's for specific chemicals. However, individual states also have the authority to set their own standards, so long as their standard is at least as stringent as the Federal standard. Thus, monitoring systems must be adjustable to allow the monitoring to be sensitive to differing MCL's in different states.
There is not known in the art a continuous, in-line, contamination monitor for heavy metals that will detect contamination spikes in a water flow, is sensitive enough to measure vastly differing concentration levels for different contaminants, is adjustable to allow the monitoring to be sensitive to differing MCL's in different states, is economical, is not wastefull of water, is capable of detecting when a break-through of a heavy metal filter has occurred, and is capable of providing a user with an alarm to designate when such a filter needs to be replaced.